PDF to Text

When you just want the words

filesmith.io/convert/pdf-to-txt

Sometimes you don’t need the layout — you need the text, to paste into an email, a search box, a translator or a script. PDF to Text strips away fonts, columns and images and gives you the readable content as a plain .txt file.

Reading order matters

A PDF stores characters by position, not by reading order, so naive extraction can jumble columns and headers. Filesmith groups characters into lines and paragraphs the way a person reads them, so the text comes out in the order you’d expect rather than scrambled.

What plain text can’t keep

Text has no concept of fonts, colours, images, headers or footers, so those are dropped by nature. What you get is the words, in order — nothing more, which is exactly the point.

Scanned PDFs won’t work

If the PDF is a scan or photo, the page is an image with no text layer to extract. That needs OCR, which Filesmith doesn’t do.

Small things that show up in the output

Extracting text is mostly clean, but a few artefacts of how PDFs store type are worth recognising so you know they aren't errors on your end:

• Ligatures. Many typefaces draw "fi" and "fl" as single joined characters. Well-made PDFs map those back to the individual letters; older or unusual ones sometimes don't, leaving odd characters mid-word.
• Hyphenated line breaks. A word split across two lines in the original may arrive still split, since the hyphen is a real character in the file.
• Tables. With no grid in plain text, columns arrive as runs of text one after another. Fine for reading, not for data.
• Headers and footers repeat on every page in the source, so they repeat in the output too.

What plain text is genuinely good for

Stripping the formatting is often the entire point:

• Searching a long document, or grepping across many at once
• Counting words in something that won't give you a word count
• Feeding content into another tool — a translator, a summariser, a script — that wants text rather than a document
• Comparing two versions of a document, since plain text diffs cleanly and PDFs don't
• Getting readable content out of a PDF whose layout fights screen readers

The output is UTF-8, so accented characters, non-Latin scripts and symbols survive intact.